Butterworth Filter Transfer Function: Circuit Design & Component Guide

A Butterworth filter transfer function is a mathematical model that describes a frequency response with a maximally flat passband and no ripples, rolling off at a predictable rate based on the filter's order. In a real circuit, this function dictates the exact resistor and capacitor ratios required to achieve a specific cutoff frequency without introducing passband ripple that could distort audio transients or sensor baselines. Beginners commonly confuse it with Chebyshev filters (which trade passband ripple for a steeper roll-off) or Bessel filters (which prioritize linear phase response and step-response fidelity over amplitude flatness).

The Mathematics of Maximum Flatness

The core promise of the Butterworth response is amplitude flatness in the passband. The magnitude squared of the transfer function for an n-th order low-pass Butterworth filter is defined as:

|H(jω)|² = 1 / [1 + (ω/ω_c)^(2n)]

Here, ω is the angular frequency, ω_c is the cutoff angular frequency, and n is the filter order. At the exact cutoff frequency (ω = ω_c), the denominator becomes 2, meaning the power is halved. This translates to an amplitude drop of -3.01 dB, regardless of the filter order. This invariant -3 dB point is a critical anchor when cascading multiple stages.

As frequency increases well beyond ω_c, the roll-off rate asymptotes to -20n dB/decade. A 2nd-order filter drops at -40 dB/decade, while a 4th-order drops at -80 dB/decade. The poles of the Butterworth transfer function are distributed evenly along the left half of the unit circle in the s-plane, which is what mathematically guarantees the absence of ripples in the passband.

Worked Numeric Example: 2nd-Order Low-Pass Sallen-Key

Let’s translate the transfer function into physical components using the Sallen-Key unity-gain topology. We will design a 2nd-order (n = 2) low-pass Butterworth filter with a target cutoff frequency (f_c) of 1 kHz.

For a unity-gain Sallen-Key Butterworth response, the damping factor (ζ) must be exactly 1/√2 (approximately 0.707). To simplify the math and component sourcing, we set the capacitors equal (C1 = C2 = C) and the resistors equal (R1 = R2 = R). The cutoff frequency equation simplifies to:

f_c = 1 / (2πRC)

Step 1: Pick a standard capacitor value.
We choose C = 10 nF. (We will address dielectric selection in the component realities section below).

Step 2: Calculate the required resistance.
R = 1 / (2π × 1000 Hz × 10 × 10^-9 F)
R = 15,915.5 Ω

Step 3: Select standard 1% (E96 series) resistors.
The closest E96 value is 15.8 kΩ. Using 15.8 kΩ and 10 nF shifts our actual cutoff frequency to 1,007 Hz, an acceptable 0.7% deviation for most analog front-ends.

Step 4: Choose the operational amplifier.
For a 1 kHz audio or sensor application, a dual op-amp like the Texas Instruments OPA2134 (FET input, low distortion) or the TL072 (budget JFET) provides more than adequate Gain-Bandwidth Product (GBW) and slew rate.

Where You Meet This in Practice

The Butterworth transfer function is not just academic; it is the default choice in several critical hardware domains:

• ADC Anti-Aliasing: When feeding signals into a 16-bit ADC like the ADS1115, you must attenuate frequencies above the Nyquist limit to prevent aliasing. A 2nd or 4th-order Butterworth filter provides a smooth passband that preserves the DC accuracy of sensor readings without introducing the passband ripple of a Chebyshev filter, which would manifest as measurement noise.
• Audio Crossovers: The Linkwitz-Riley crossover, standard in professional loudspeaker management, is literally two cascaded Butterworth filters. This specific configuration yields a -6 dB point at the crossover frequency, ensuring the acoustic sum of the woofer and tweeter remains perfectly flat.
• Biomedical Signal Conditioning: In ECG and EEG front-ends, baseline wander and high-frequency EMG noise must be stripped away. Butterworth bandpass filters are preferred because the maximally flat passband ensures the morphology of the QRS complex or brainwave spikes is not amplitude-distorted.

Decision Path: Topology and Order Selection

Choosing the right filter order and topology prevents over-engineering while guaranteeing signal integrity. Use this decision matrix to terminate your design choices.

Component Realities: What the Transfer Function Ignores

The mathematical transfer function assumes ideal, infinite-impedance, zero-tolerance components. On the workbench, parasitics and material physics will ruin your Butterworth response if you ignore them.

Furthermore, the operational amplifier's Gain-Bandwidth Product (GBW) must be vastly larger than the filter's cutoff frequency. A common rule of thumb is that the op-amp's GBW should be at least 100 × f_c × Q, where Q is the quality factor of the specific filter stage. For a 2nd-order Butterworth stage, Q is 0.707. If your f_c is 10 kHz, you need an op-amp with a minimum GBW of 100 × 10,000 × 0.707 = 707 kHz. While a cheap LM358 (GBW ~1 MHz) might barely pass this, its poor slew rate and crossover distortion will inject harmonics. Step up to an OPA2277 (GBW 1 MHz, but precision specs) or an NE5532 (GBW 10 MHz) for a robust margin.

For automated component calculation and parasitic modeling, utilize the Analog Devices ADIsimFilter tool or the Texas Instruments Filter Designer. These tools automatically flag when your selected op-amp lacks the GBW to support your target transfer function.

Butterworth Filter FAQs